Simple datastore for arbitrary data, fronted by grafana. While Grafana fetches the data normally, this makes it easy for applications to push their own data. Allows setting a custom 'received time' to either push old data or later push data when internet might be available.

Both simple values and complex JSON data is accepted, if the <APP_NAME> doesn't exist already, it is created.

curl -X POST -H "Content-Type: text/plain" --data "<DATA>" <URL>/app/<APP_NAME>

Note: Optionally ?created=<UNIX_TIME> can store the entry as received at certain time.

./rb.sh (runs setup as docker-compose)
./seed_dms.sh (for some initial test data)

Navigate to http://0.0.0.0:3000, follow Grafana setup. (The underlaying API serves from http://0.0.0.0:5000)

Use the SimpleJson datasource as follows:

Because complex data isn't available as timeseries normally, a property can be chosen to represent the table-data when shown as timeseries:

curl -X POST -H "Content-Type: text/plain" --data '{"data_series_var": "<PROPERTY_PATH>"}' <URL>/app/<APP_NAME>/config

The data_series_var property describes the path to the desired property:

{
    "my": {
        "nested": {
            "property": 42
        }
    }
}

{"data_series_var": "my.nested.property"}

allowing fancy timeseries-graphs from pushed JSON

From here, data that is pushed under new app-names automatically appear as metrics.

• Creates app on first push

curl -X POST -H "Content-Type: text/plain" --data "<DATA>" <URL>/app/<APP_NAME>

Configure app

curl -X POST -H "Content-Type: text/plain" --data '{"data_series_var": "<PROPERTY_PATH>"}' <URL>/app/<APP_NAME>/config

Delete app

curl -X DELETE <URL>/app/<APP_NAME>

Run tests (requires python dependencies available):

./test.sh $(<project_root_path>)